Tianxue Zhang

Tianxue Zhang

Auto-generated profile, based on publicly available information

Tianxue Zhang

School of Mechanical Engineering and Automation, Beihang University, Beijing, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 December 2024 in Agriculture
Verified
Journal Article
Open access
PDF file
ALAo Li
Ao Li
CWChunrui Wang
Chunrui Wang
TJTongtong Ji
Tongtong Ji
QWQiyang Wang
Qiyang Wang
TZTianxue Zhang
Tianxue Zhang
Accurate and efficient tomato detection is one of the key techniques for intelligent automatic picking in the area of precision agriculture. However, under the facility scenario, existing detection algorithms still have challenging problems such as weak feature extraction ability for occlusion conditions and different fruit sizes, low accuracy on edge location, and heavy model parameters. To address these problems, this paper proposed D3-YOLOv10, a lightweight YOLOv10-based detection framework. Initially, a compact dynamic faster network (DyFasterNet) was developed, where multiple adaptive convolution kernels are aggregated to extract local effective features for fruit size adaption. Additionally, the deformable large kernel attention mechanism (D-LKA) was designed for the terminal phase of the neck network by adaptively adjusting the receptive field to focus on irregular tomato deformations and occlusions. Then, to further improve detection boundary accuracy and convergence, a dynamic FM-WIoU regression loss with a scaling factor was proposed. Finally, a knowledge distillation scheme using semantic frequency prompts was developed to optimize the model for lightweight deployment in practical applications. We evaluated the proposed framework using a self-made tomato dataset and designed a two-stage category balancing method based on diffusion models to address the sample class-imbalanced issue. The experimental results demonstrated that the D3-YOLOv10 model achieved an mAP0.5 of 91.8%, with a substantial reduction of 54.0% in parameters and 64.9% in FLOPs, compared to the benchmark model. Meanwhile, the detection speed of 80.1 FPS more effectively meets the demand for real-time tomato detection. This study can effectively contribute to the advancement of smart agriculture research on the detection of fruit targets.

25 August 2022 in IEEE Transactions on Medical Robotics and Bionics
Verified
Journal Article
Closed access
ZLZecai Lin
Zecai Lin
TZTianxue Zhang
Tianxue Zhang
ZSZhenglong Sun
Zhenglong Sun
HGHongyan Gao
Hongyan Gao
XAXiaojie Ai
Xiaojie Ai
AGAnzhu Gao
Anzhu Gao
Robotic telemedicine has a high potential to promote the development of future medicine. It enables the treatment of critical patients in geographically remote locations in time, and also particularly useful for avoiding infection risks during pandemics of infectious diseases. However, visual feedback from the patient’s side to the clinician’s side is often unintuitive due to a lack of depth information, occlusion, and viewing angles. This paper develops a robotic telepresence method based on augmented reality and human motion mapping for interventional medicine. An optical see-through head-mounted display (OST-HMD) is mounted at the distal end of a redundant manipulator, which can provide views from different directions, and the clinician can see through key anatomies and instruments located inside the patient body. The control of the manipulator is accomplished via human motion mapping, with both the end-effector and manifold of the human arm being mapped to the manipulator. A simple evaluation method is developed to assess the accuracy of head localization of the OST-HMD. Telepresence experiments are carried out to verify the effectiveness of the proposed method, with which the operator can see through a human airway phantom.

4 December 2019 in IEEE Transactions on Medical Robotics and Bionics
Verified
Journal Article
Closed access
TZTianxue Zhang
Tianxue Zhang
Yongshuang WenYongshuang Wen
Yongshuang Wen
YLYunhui Liu
Yunhui Liu

22 January 2026 in Agriculture
Verified
Journal Article
Open access
PDF file
ALAo Li
Ao Li
CWChunrui Wang
Chunrui Wang
AWAichen Wang
Aichen Wang
JSJianpeng Sun
Jianpeng Sun
FGFengwei Gu
Fengwei Gu
TZTianxue Zhang
Tianxue Zhang
Segmentation of tomato fruits under complex lighting conditions remains technically challenging, especially in low illumination or overexposure, where RGB-only methods often suffer from blurred boundaries and missed small or occluded instances, and simple multimodal fusion cannot fully exploit complementary cues. To address these gaps, we propose YOLO-MSRF, a lightweight RGB–NIR multimodal segmentation and refinement framework for robust tomato perception in facility agriculture. Firstly, we propose a dual-branch multimodal backbone, introduce Cross-Modality Difference Complement Fusion (C-MDCF) for difference-based complementary RGB–NIR fusion, and design C2f-DCB to reduce computation while strengthening feature extraction. Furthermore, we develop a cross-scale attention fusion network and introduce the proposed MS-CPAM to jointly model multi-scale channel and position cues, strengthening fine-grained detail representation and spatial context aggregation for small and occluded tomatoes. Finally, we design the Multi-Scale Fusion and Semantic Refinement Network, MSF-SRNet, which combines the Scale-Concatenate Fusion Module (Scale-Concat) fusion with SDI-based cross-layer detail injection to progressively align and refine multi-scale features, improving representation quality and segmentation accuracy. Extensive experiments show that YOLO-MSRF achieves substantial gains under weak and low-light conditions, where RGB-only models are most prone to boundary degradation and missed instances, and it still delivers consistent improvements on the mixed four-light validation set, increasing mAP0.5 by 2.3 points, mAP0.5–0.95 by 2.4 points, and mIoU by 3.60 points while maintaining real-time inference at 105.07 FPS. The proposed system further supports counting, size estimation, and maturity analysis of harvestable tomatoes, and can be integrated with depth sensing and yield estimation to enable real-time yield prediction in practical greenhouse operations.

24 April 2026 in Nature Communications
Verified
Journal Article
Open access
PDF file
HLHaiming Li
Haiming Li
PGPeiyuan Gao
Peiyuan Gao
HTHaoyue Tan
Haoyue Tan
HLHaotian Li
Haotian Li
HZHaifeng Zhu
Haifeng Zhu
JZJie Zhao
Jie Zhao
Intracochlear theranostics, particularly targeted drug delivery and microsampling, offers a promising solution to inner ear diseases. However, specialized medical devices remain limited by a fundamental design challenge imposed by anatomical constraints: balancing miniaturization, dexterity, and perceptive functionality. Here, we present a low–aspect-ratio, dual-segment continuum robot that integrates catheter, endoscopic and instrumental functions. Driven by antagonistic cables, the robot uses a transition-free backbone composed of saddle-shaped joints to achieve a minimum bending radius of 1.9 mm. Dual-segment motion decoupling yields programmable C-/S-shaped configurations, facilitating anatomical navigation. The microneedle, embedded in the central channel, serves as an end-effector with positioning accuracy of 17.9 ± 4.1 μm. Fiber Bragg grating sensors, mounted on the needle, measure axial force to estimate tool-tissue interaction. Validation is performed on cadavers and in vivo animals, demonstrating the feasibility of a transcanal, atraumatic robotic paradigm. Thus, this system provides a practical and accessible approach for early diagnosis and treatment, helping extend precision medicine to underserved areas.

13 December 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
TZTianxue Zhang
Tianxue Zhang
AGAnzhu Gao
Anzhu Gao
JZJianjun Zhu
Jianjun Zhu
BZBohan Zhang
Bohan Zhang
JLJiachun Lin
Jiachun Lin
YNYuting Ni
Yuting Ni
Fiber Bragg grating (FBG) optical fiber shape sensing has the potential to minimize harmful radiation exposure for both medical professionals and patients. It provides real-time, three-dimensional visualization of vascular structures during interventional procedures. The field has witnessed advancements through the development of a multi-core 5-FBG sensor array engineered to capture critical attributes inherent to vascular morphology, such as curvature and torsion. An optimized helical extension method (HEM) has been introduced to accurately measure vessel curvature degrees and facilitate precise three-dimensional reconstruction of vascular shapes. This refined HEM algorithm has significantly reduced the reconstruction error rate to 15.611%. An experimental system for interventional vascular optical fiber sensing was established to validate the effectiveness of the proposed method. The obtained vascular shape measurements were compared with those derived from computed tomography (CT) imaging. The maximum error between the two measurements was found to be less than 1.4156 mm, affirming the accuracy of the enhanced HEM algorithm in perceiving vascular shape. This research demonstrates the feasibility of utilizing the constructed multi-core 5-FBGs optical fiber for shape perception in a wide range of vascular structures. This study serves as a foundation for the integration of FBG sensors into clinical vascular shape perception approaches, presenting the potential benefits of reduced radiation exposure and improved accuracy in interventional procedures.